United States

(1)

Data from the

NATJC)NAL HEALTH SURVEY

Skeletal Maturity of Children

641 ^years:

Racial, Geographic Area,

and Socioeconomic Differentials

United States

Skeletal age (hand-wrist) and bone-specific skeletal ages (as determined using the Health Examination Survey standard based on the Greulich-P yle Radiographic Atlas) of boys and girls 6-11 years of age, by chronological age, race, geographic region, population size or land use of area of residence, annual family income, and education of parent.

DHEW Publication No. (HRA) 76-1631

U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE Public Health Service

Series 11 Number 149

Health Resources Administration National Center for Health Statistics Rockville, Md. October 1975

(2)

Library of CongressCataloging in Publication Data Roche, Alexander F.

Skeletal maturity of children 6-11 years: Socioeconomic differentials.

(National Center for Health Statistics. Data from National Health Survey, series 11, no.

149) (DHEW publication no. (HRA) 76-1631) Includes bibliographical references.

Sqpt. of Dots. no.: HE 20.2210:11/149

1. Children-Growth-Testing. 2. Bone–Growth–Statistics. 3. Children in the United States–Anthropometry. I. Roberts, Jean, joint author. II. Hamill, Peter V. V., joint author.

III. Title. IV. Series: United States. National Center for Health Statistics. Vital and health statistics. Series 11: Data from the National Health Survey. Data from the health examination survey, no. 149. V. Series: United States. Department of Health, Education, and Welfare. DHEW publication no. (HRA) 76-1631. [DNLM: 1. Age determination by skeleton. 2. Bone development. 3. Hand—Growth & development. 4. Socioeconomic factors.

W2 A N148vk no. 149]

M407.3.A347 no. 149 [RJ13] 312’.0973s [312’.6]

ISBN 0-8406-0030-5 75-18049

(3)

NATIONAL CENTER FOR HEALTH STATISTICS

HAROLD MARGULIES, M.D., Acting Director

ROBERT A. ISRAEL, Acting .Deputy Director

GAIL F. FISHER, Associate Director for the Cooperative Heultk Statistics System ELIJAH L. WHITB, Associate Director for Data Systems

EDWARD E. MINTY, Associate Director for Management PETER L. HURLEY, Acting Associate Director for operations JAMES M. ROBEY, Ph.D., Associizte Director for Program Development

ALICE HAYWOOD, Information Oflicer

DIVISION OF HEALTH EXAMINATION STATISTICS

ARTHUR J. McDOWELL, Director

JEAN-PIERRE HABICHT, M.D., Ph.D., Special Assistant to Director PETER V. V. HAMILL, M.D.,Medical Adviser

JEAN ROBERTS, Chiej Medical Statistics Brarzch

ROBERT S. MURPHY, Chief Suruey Planning and Development Branch

COOPERATION OF THE BUREAU OF THE CENSUS

In accordance with specifications established by the National Center for Health Statistics, the Bureau of the Census, under a contractual agreement, participated in the design and selection of the sample, and carried out the first stage of the field interviewing and certain parts of the statistical processing.

Vital and Health Statistics-Series 1 l-No. 149

DH EW Publication No. (H RA) 76-1631

(4)

0.0 *

(6)

SKELETAL MATURITY

iDF CHILDREN 6-11 YEARS:

RACIAL, GEOGRAPHIC AREA, AND SOCIOECONOMIC DIFFERENTIALS

A. F. Roche, M.D., Fels Research Institute, Jean Roberts, and Peter V. V. Hamill, M.D.,

Division of Health Examination Statistics

INTRODUCTION

This report contains national estimates of the levels of skeletal maturity in the hand-wrist of noninstitutionalized children in the United States 6-11 years of chronological age by race, area of residence, and socioeconomic back- ground, as determined in the Health Examina- tion Survey of 1963-65. The previous report on these radiographic findings,l which was limited to the sex differences in skeletal maturity in relation to chronological age, contains a more comprehensive description of the nature of skeletal maturation and of the assessment method used in the Survey.

The Health Examination Survey is one of the major programs of the National Center for Health Statistics, authorized under the National Health Survey Act of 1956 by the 84th Congress as a continuing Public Health Service function to determine the health status of the population.

In carrying out the intent of the National Health Survey,z four different types of survey programs are used. The Health Interview Survey collects health information from samples of people by household interview with a responsi- ble adult and is focused primarily on the impact of illness and disability within various population groups. The programs in the Divisions of Health Resources Utilization Statistics and Health Manpower and Facility Statistics obtain health data as well as health resource and

utilization information through surveys of hos- pitals, nursing homes, and other resident institu- tions, and also data on the entire range of personnel in health-related occupations. The Health Examination Survey collects health data by direct physical examination, tests, and measurements performed on samples of the population. The latter program provides the best way of obtaining the actual prevalence of certain medically defined illnesses. It is the only effec- tive way to secure information on unrecognized and undiagnosed conditions and on a variety of physical, physiological, and psychological measures within the population. Also it collects demographic and socioeconomic data on the sample population under study to which the examination findings may be related.

The Health Examination Survey (HES) is organized as a series of separate programs, or cycles, each of which is limited to some specific segment of the United States population and to specific aspects of health. In the first cycle, the prevalence of certain chronic diseases and the distribution of various physical and physiological measures were determined on a cross section of the defined adult population, as previously described.3 *4

In “;he program on which the findings in this repm-twm+ased, a probability sample representative of the approximately 24 million noninstitutionalized children aged 6-11 years in the United States was selected and

(7)

examined. The examination in this cross- sectional study primarily assessed health factors related to growth and development. It included an examination by a pediatrician and by a dentist, tests administered by a psychologist, and a variety of tests and measurements administered by a technician. A more complete description of the survey plans, sample design, examination content, and operation of the Survey has been published.5

Field collection operations for this cycle were started in July 1963 and completed in December 1965. There were 7,417 children selected in the sample, of whom 7,119, or 96 percent, were examined. This national probability sample is representative, and those examined are closely representative of the approximately 24 million noninstitutionalized children aged 6-11 years in the United States with respect to age, sex, race, region, size of place of residence, and rate of change in size of place of residence from 1950 to 1960. Of the children examined, 157 were not radiographer or had radiographs taken for the Survey that were not suitable for assessment.

Thus, the skeletal maturity estimates for United States children 6-11 years are based on the remaining 6,962 children, or 94 percent of the original sample.

The examining team gave each child a standardized examination during hk single visit in the mobile units specially designed for use in the Survey. Demographic and socioeconomic data on household members as well as medical history, behavioral, and related data on the child to be examined were obtained from his parents prior to his examination. Ancillary data were re- quested from the school attended by the child, including his grade placement, school behavior, adjustment, and health problems: Birth certifi- cates for verification of the child’s age and information related to his condition at birth were obtained ajso.

The present analysis concerns differences in skeIetal maturity levels for the United States population aged 6-11 years, grouped according to race, region, and socioeconomic status. The general concept of skeletal maturity, the method by which radiographs were taken and later assessed, and the quality control measures used have been described and discussed.1 The present findings provide normative national data. They

are interpreted in regard to the health signifi- cance of skeletal maturity status and compared with findings from other studies. Not all relevant reports have been included in the literature review, some because the sample size was inadequate but others because of inadequate docu- mentation concerning the repeatability and comparability of the skeletal age assessments. In later reports, the skeletal ages of these children will be considered in relation to body size, body composition, and other variables.

A brief description of the sample design, quality control methods, reliability of the data, as well as population and sampling error estima- tion procedures used for the findings of this study is contained in appendix I. Definitions of the demographic terms used in the report are included in appendix II, and an evaluation of the reliabilityy of the assessments is shown in appendix III.

EXAMINATION METHOD

At each of the 40 preselected locations throughout the United States used consecutively in this cross-sectional study, the children were brought to the centrally located mobile examination center for a standardized examination that lasted about 2?42hours. Six children were examined in the morning and six in the after- noon. When each child entered the mobile center, his oral temperature was taken and a screening for acute illness was made. If such illness were detected, the child was sent home and reexamined later. Each examinee next dressed in shorts, cotton sweat socks, and a light sleeveless shirt and proceeded to a designated but different station for the examination. The sequence of elements in the examination dif- fered for each child, so that the six could be examined simultaneously during the half day.

The same examiners–physician, dentist, psychologist, and specially trained technicians—

conducted their parts of the examination in essentially the same manner for each child. The time of each part of the examination was recorded, but there is no reason to believe that diurnal or sequence effects would be present in the composition or quality of the radiographic data.

(8)

Field Radiography

Each child was scheduled to have a 10” X 12”

radiograph taken of the right hand and wrist for which the positioning was otherwise in accordance with specifications in the GreuIich-Pyle Atlas.G Some radiographs were made using other film sizes when the 10“ X 12“ size was scarce;

this would not have influenced the findings.

Technically inadequate films could be repeated because they were developed immediately in the field. Thus, each child’s record contains a single radiograph showing the dorsopalmar view of his entire hand-wrist with its full complement of ossifying parts, at his examination age.

The decision to radiograph the right hand- wrist rather than the left, which is the more frequent anthropometric practice, was made on the advice of anthropologist consultants who were interested also in the use of related measurement data for equipment design in which right-side measurements were preferred.

When the selected plates from GreuIich-Pyle Atlas standards and those from other sources were reproduced in the HES Standard, they were reversed photographically so that they could be used in right-side assessments. Previous reported research7 on lateral differences in the skeletal maturity of the hand-wrist, either for the area as a whole or bone by bone, has shown that these are too small to be of practical importance.

Training of Assessors

The assessment of skeletal age from the hand-wrist radiographs of children 6-11 years of age in the Health Examination Survey of 1963-65 was made by six medical students at Case Western Reserve University, one of whom was an instructor specializing in anatomy. This work was done under contract for the National Center for Health Statistics, with Dr. P. Wesley Dupertius as Project Director. Training of the assessors and implementing the related quality control procedures were done in the meticulous manner previously described under the direc- tion of Dr. S. I. Pyle. When the ratings and reliability for the new assessor were in good agreement with those of Dr. Pyle (the majority of differences within 4 months) the new assessor

started his assessments of the survey radiographs. Reported evidences suggests that, at the end of the training procedure, the interobserver and intraobserver differences in skeletal maturity ratings were similar to those for experienced assessors.

Assessment Pr6cedure

The radiographs were assessed by comparison with prints of the series of standards for the male hand-wrist selected from those in the 1959 Greulich-Pyle AtlasG and other sources. These were reversed so that they appeared to be of the right hand-wrist as shown in the 1971 Radio- graphic Standard of Reference of Pyle, Water-

house, and Greulich.g This standard contains the male skeletal age equivalents that were used during the assessment of Cycle II radiographs (ages 6-11 years), with some very slight modifi- cation to smooth the skeletal age trend for a few of the bones.

The readers did not have access to the chronological age, sex, or other information about the chiId. Each bone was assessed separ- ately and interpolation was made between the standards to monthly intervals when this appeared appropriate.

As a quality control measure and to permit determination of the level of reliability of the assessments throughout this study, independent replicate assessments were obtained on approximately 1 out of each 11 films. One randomly selected radiograph in each 23 was rated inde- pendently by another assessor for a measure of interobserver variability and 1 randomly selected radiograph among each 20 was rated independ- ently a second time by the same reader to give a measure of intraobserver variability. The time lapse between the first rating and the reassessment was sufficiently long that there was little likelihood of recall. Furthermore, there was no indication to the assessor that he was making a reassessment. Information on the degree of reliability of assessments from this survey has been published.1

The order in which the bones were assessed within each radiograph was the same as that in which they are listed in table A. This table ako shows the minimum and maximum individual bone-specific skeletal ages allowable in this

(9)

Table A. Minimum and maximum acceptable skeletal ages (in months) using the HES Standard

Hand-wrist bone

Radius . . . . Ulna . . . . Cavitate . . . . Hamate . . . . Triquetral . . . . Lunate . . . . Scaphoid . . . . Trapezium . . . . Trapezoid . . . . Metacarpal I . . . . Metacarpal lo . . . . Metacarpal ill . . . . Metacarpal IV . . . . Metacarpal V . . . . Proximalphalanxl . . . . Proximal phalanxll-V . . . . Middle phalanx ll-lV . . . . Middle phalanx V . . . . Distal phalanxl . . . . Distal phalanx II, V . . . . Distal phalanx Ill . . . . Distal phalanx IV . . . .

Pisiform . . . . Adductor sesamoid . . . . Flexor sesamoid . . . .

IEzEizz

Skeletal age in months 15

69 -- . ..

17 35 68 51 68 25 17 16 17 24 33 15 23 39 15 39 22 32 110 146 158

228 215 197 197 197 197 197 197 197 191 215 209 209 215 215 209 209 209 191 191 191 191 197 197 197

1 Minimum age (according to standard) of tha radio-opacity of the epiphysis or carpal.

2 One month below “adult” age.

study. The lower limit for the bone-specific skeletal age was arbitrarily set midway in skeletal age (hand-wrist) between the last standard in which the particular bone was not radio- opaque and the first in which it was radio- opaque. Exceptions were made for the three later ossifying bones–the pisiform and the adductor and flexor sesamoids–for which minimum ages allowed were 2 months above the last standard in which they were not radio-opaque.

There are limits also at the upper end of the range when bones become adult. Only the designation “adult,” and not a skeletal age in months, can be assigned to a bone in which maturation is complete. Median ages in months from the HES Standard at which this occurs in boys were used as the skeletal age for each bone

beyond which only the designation “adult”

could be applied. The maximum allowable values 1 month below the “adult” skeletal age are shown in table A.

As expected, withih chronological age groups, the skeletal ages assigned to girls were more advanced than those assigned to boys. This occurs because, although boys and girls pass through the same skeletal maturity stages, girls tend to mature more rapidly than boys. The female equivalent skeletal ages were determined during the preparation of the HES Standard, but were not used in the assessment of the Survey radiographs. The method by which these female equivalent skeletal ages were obtained has been described in detail by Pyle et al.g

The skeletal age data for girls in the detailed tables of this report are given both in terms of the male standards, as originally assessed, and in terms of the female equivalent skeletal ages.

However, the findings for girls in the text are limited to the female equivalent skeletal ages.

The skeletal age values for the whole hand-wrist for boys and girls in this report were determined by computer from the original bone-specific assessments by averaging the ages assigned the hand-wrist bones for each child.

SKELETAL AGE (HAND-WRIST) FINDINGS

Among both boys and girls 6-11 years of age in the United States (considering only the female equivalent values for girls), skeletal age (hand-wrist) was found, on the average, to be consistently less than chronological age. The mean sex-associated differences increase steadily with chronological age. Comparison of skeletal age for boys with the female equivalent values for girls shows close agreement (mean difference less than 1 month) among children 7 and 8 years of age at their last birthday, while among younger and older children of 6 and 9-11 years, the meris for boys are substantially less advanced than those for girls, by 2 to 3 months.

Boys showed no consistent pattern in the variability of skeletal age (hand-wrist) with chronological age, but older girls were slightly more variable than younger girls in this respect.

Except at ages 9 and 10 years, boys showed

(10)

slightly greater relative variability than girls in their skeletal ages. These national estimates, which have been reported previously, 1 are based on data from the National Health Examination Survey of 1963-65 in which assessments of hand-wrist radiographs were made using the HES Standard based primarily on the Greulich-Pyle Radiographic Atlas.b

able environmental factors—e.g., protein intake, temperature–on skeletal maturation in children.

This report contains information on the racial, geographic region, urban-rural, and socioeconomic differentials in the skeletal maturity of United States children. These findings are compared with relevant research findings from previous studies.

Pre;io& international research has shown “ clearly that children of different racial groups Race vary in body size and in levels of skeletal and

sexual maturity at corresponding chronological ages,lo‘15 Factorsinfluencing the rate of the apposition or resorption of bone in skeletal maturation are related to genes, nutrition, illness, or climate. Only in a few studies, however, have the possible effects of these factors been separated. Interactions between these factors make it difficult to draw definitive conclusions from reported findings outside the context of animal experimentation. Knowledge is almost entirely lacking concerning the effects of specific groups of genes—e.g., autosomal, sex- Iinked–or the effects of quantitative, measur-

The skeletal maturity of white boys and girls 6-11 years of age generally tends to be somewhat less advanced than that of Negro boys and girls in the United States of corresponding chronological age. The number of children of other races in this country, and hence in the study sample, is too small to provide reliable estimates for this heterogeneous segment of the population.

Among white boys, the mean skeletal age (hand-wrist) is consistently lower than that of their Negro counterparts, except for those 10 years of age at their last birthday (figure 1 and

BOYS

6 7 8 9 10 11

CHRONOLOGICAL AGE IN YEARS

r

^GIRLS

5 (Female equivalents)

I

Negro

White –15

t

I Î Î Î Î Î Î

6 7 8 9 10 11

CHRONOLOGICAL AGE IN YEARS

Figure 1, Mean difference in months between skeletal age (hand-wrist) and chronological a9e for white and Negro boys and 9irk by chronological age in years: United States, 1963-65.

(11)

table 1). At all but 6 and 8 years (chronological age) where the skeletaI age means for Negro boys exceed those for white boys by more than 4 months, the mean differences between the races in skeletal age are small (1 to 2 months).

With the size and design of the sample used in this study, they could easily reflect sampling variabilityy alone (i.e., the differences are not statistically significant at the 5-percent probability level using standard parametric tests—see appendix I). These differences are also not significant when data for all ages are combined.

The mean skeletal age (hand-wrist) of white girls (female equivalent values) also lags behind that for their Negro counterparts, except for those 8 years of age at their last birthday where the mean values are identical. For the remainder of the girls, mean differences vary between O.7 month at chronological age 11 years and 2.7 months at age 10 years. All of these differences are within the 95-percent confidence limit for such estimates and hence could reflect sampling variability alone. When data are combined for 6-11 years, these white-Negro differences are statistically significant for boys but not girk.

When indirect adjustments are made for any differences in the age distributions among white and Negro children (assuming that the national age-specific mean values apply identically in both racial groups), the contrast in skeletal maturity bet we en the two racial groups is even more clear. This is shown for all ages combined in figure 2.

5!-

-5 !-

WHITE NEGRO

Figure 2. Difference between actual and axpected mean skeletal age (hand-wrist) for whita and Negro boys and girls 6-11 years of chronological age: United States, 1963-65.

Among white children, girls are genedly more advanced than boys in their skeletal age (hand-wrist) except at chronological age 8 years when the mean values are identical (figure 3).

The differences in mean skeletal age for the remainder range from 1.2 months at age 7 years to 4.3 months among the oIdest children at age 11 years. Among the youngest and oldest children—those of chronological ages 6, 10, and 11 years–the mean differences exceed the 95-percent confidence limit for such estimates and hence are not likely to be due to sampling variabilityy alone.

Negro girls of chronological ages 9-11 years are also generally more advanced in skeletal age than Negro boys by mean values of 2.9, 7.0, and 3.0 months, respectively. The differences at 9 and 11 years are too small to be considered statistically significant (at the 5-percent probability level). Among younger Negro children of 6 and 7 years (chronological age), the mean differences in skeletal age between b oys and girls are negligible, but among those 8 years of age at their last birthday, Negro boys were more advanced than Negro girls in skeletal age (female equivalent values) by more than 4 months.

The variation in skeletal age (hand-wrist) among children of chronological ages 6-11 years in this country shows no consistent pattern of racial differences. White boys of chronological ages 6, 8, and 9 years are relatively more variable in skeletal age than Negro boys while at ages 7, 10, and 11 years the reverse was found (figure 4 and table B). White girls at alI ages except 7 and 10 years, where the differences were small, are somewhat less variable in this respect than Negro girls. The pattern of relative variabilityy between boys and girls differs for white and Negro children. Among white children, boys showed greater relative variability in skeletal age than girls in four of the six age groups-6-8 and 11 years–while among Negro children, boys were Iess variable than girk in skeletal age at aII but 7 years of age. The patterns of sex differences in variability, across age, were more regular in white than in Negro children. This greater stability in the white children probably reflects the greater sample size for them.

O~her studie; (white). –Comparison with racial findings

children

for skeletal maturity- of United States from the present national survey is

(12)

WHITE

5

0

%..●✎✎✎✎

●☛

–5

–lo (Female equivalents)

–15

;~

6 7 8 9 10 11

CHRONOLOGICAL AGE IN YEARS

It t Î ⁱ Î ¹ Î

6 7 8 9 10 11

Figure3, Mean difference in months between skeletal age (hand-wrist) andchronologiml age forwhite boys andgirls and Negro boys andgirls, by chronological age in years: United States, 1963-65.

20

15

10

5

0

BOYS GIRLS

....

‘.*

“.‘.

t I I I 1 I

6 7 8 9 10 11

20

15

10

5

0

-’ ^z

6 7 8 9 10 11

Figure4, Relative variability in skeletal age (hand-wrist) among white and Negro boys andgirls, bychronological ageinyears: United States, 1963-65.

(13)

Table B. Coefficients of variability of skeletal agewithinchron- ological age groups

Chronological aga

6years . . . . 7years . . . . 8years . . . . 9years . . . . 10yaars . . . . Ilyaars . . . .

Boys I Girls

1

White Negro White Negro

Coefficient of variation (100 sx/X) 15.3

13.4 12.4 10.8 9.0 10.5

13.6 14.7 10.0 10.0 9.2 10.9

13.9 12.0 10.8 11.7 12.6 9.4

14.6 11.5 12.8 14.6 12.5 11.0

limited primarily to previous studies among various subgroups of children in this country.

The skeletal maturity (hand-wrist) data for white children from seven previous studies in this country (some longitudinal and some at a single point in time) and the United States estimates for 1963-65 from the present national cross-sectional study are shown in figures 5-8.

Data from the previous studies have been ad- justed to a common Greulich-Pyle zero line to facilitate comparisons among them and with the present study. Data from the previous studies relate to mean, median, or approximate modal values of skeletal age (hand-wrist) within chronological age groups and are shown as the difference between skeletal and chronological

1.0

r

^— Flory (1936) --- Todd (1937) .. . . Simmons (1944)

...-- ...

....”

. . . . . . . . .

...

A

““””

.

^...-”” _/\^/\ ^,/’, _\“ _{# -,}

.~ \ /’ \

--- \,/ \.

.,.5 ~

6 7 8 9 10 ¹¹

CHRONOLOGICAL AGE IN YEARS

Figure 5. Differences between skelatal and chronological ages of boys by chronological age in years, in studies of Flory (1936), Todd (1937), and Simmons (1944).

--- Greulich and Pyle (1959) and Bayley (1962), 1.0,

[

--- Johnston (1962) --- Fry (1966)

--. Maresh (1971) /-. . . ...-’- 0.5 - — United Statea children /.

(1963-65) . . . . .. .

#.... - . . . . /

0 - ...”.

/= /’

----=. ,-___e --

: [:*

-1.5 I 1 1 1 1 r

6 7 8 9 10 11

Figure 6. Differences between skeletal and chronological ages of boys, by chronological aga in years, in studies of Greulich and Pyle (1959), Bayley (1962), Johnston (1962), Fry (1966), and Nfarash (1971), and for United States children (1963-65).

— Flory (1936) --- Todd (1 937) .. . . . Simmons (1944)

. . . . . . . .

0 -

-%.. . . .

#... - . . . .

. . . .

< /.

g /--.,~.

.. .. .

u /

~ .. .. . ---- /

–0.5 - .... *tiO----=-

i Y

./

-1.5 Ill

iY WI

-2.0

I

-2.56-

Figure 7. Differences between skelatal and chronological ages of girls (female equivalent values for the former), by chrono- logical age in years, in studies of Flory (1936), Todd (1937), and Simmons (1944).

age within chronological age groups in these figures. Values above the zero line show accelera- tion in skeletal maturity (skeletal age exceeds chronological age) and those below the zero line show retarded skeletal age. The 1959 Greulich- Pyle standards6 were selected from radiographs of white children of upper socioeconomic status

(14)

(Female equivalent values) --- Greulich and Pyle (1959) 1.0- . . . . . Bayley (1962)

~ . . . . Johnston f 1962)

G --- Fry (1966)

~ --- Maresh (1971)

0.2UI — United States children (1963-6E

Occ ma=W

u> o

i~~ ‘;’*

,. . . ..=~77k.~5F+~

w .

-1.0

6 7 6 9 10 11

Figure 8. Differences between skeletal and chronological ages of girls (female equivalent values for the former) by chronologi- cal age in years, in studies of Greulich and Pyle (1959), Bayley (1962), Johnston (1962), Fry (1966), and Maresh (1971 ), and for United States children (1963-65).

living in Cleveland. These children were born between 1917 and 1942 and were radiographed close to birthdays and half-birthdays, as previously described.1 The method by which these standard plates were selected from the 100 radiographs available for each sex at each age is described in the Atlas of Greulich and PyIe.6

The mixed longitudinal data of FIoryl 6 were obtained from white Chicago children of above- average socioeconomic status who were born between 1911 and 1923 and were radiographed close to each birthday. Flory selected from the 100 radiogmphs available for each sex at each age (except at 6 and 7 years when at least 80 radiographs were available) those radiographs that he considered best represented the central tendencies of skeletal maturity in his groups.

The selected radiographs are about 1 year retarded in maturity in comparison with corresponding GreuIich-PyIe standards (figures 5 and 7). Simmonsl 7 reported data from Cleveland children, most of whom were included in the sample studied by Greulich and PyleG and who had been radiographed near each birthday. The sample size varied from 154 to 206 for each year of age for each sex. The means reported by Simmons, after adjustment for the use of Todd standards,l S were within 0.5 year (skeletal age) of the Greulich-Pyle standards.

Toddl s published standards white Cleveland children of all

derived from socioeconomic

levels. These chiIdren were born between 1920 and 1930 and radiographer near each birthday and half-birthday. The sample size within each age-sex group varied from 35 to 94 children. The standard plates were chosen to represent the central tendencies for skeletal maturity Ievel within these age-sex-specific groups. The Todd standards (from all socioeconomic groups) were at lower levels than those of Greulich and Pyle (from upper socioeconomic groups). The Todd values were also lower than the mean skeletal maturity levels reported by Simmons (from upper socioeconomic groups) but exceeded those of Flory (from average socioeconomic groups) .

Bayleyl 9 reported data from the Harvard Growth Study of white children of middle socioeconomic status, living in Boston, born between 1930 and 1939, and examined near each birthday. The sample size varied from 63 to 67 in each age-sex group. The mean skeIetal ages (Greulich-Pyle) for these children were very c 10 se to their mean chronological ages.

Johnstonz 0 reported mixed longitudinal data for middle-class white Philadelphia children born between 1937 and 1955. The ethnic strains in this sample were (in order of frequency) Italian, Scotch, Irish, English, Polish, Russian, and Ukranian, with many Jews included among the East Central European peoples.21 These children were radiographed at random chronological ages, and the sample size varied from 23 to 51 for each annual interval in each sex. All mean skeletal ages were within O.5 year of the Greulich-PyIe standards except in 10- and 11- year-old boys who were about 0.6 to O.7 year ahead of the Greulich-Pyle standards (figure 6).

It would be tempting to conclude that the skeletal advancement of these boys reflects their ethnic origins, but this factor should then have operated equally in girls which it did not (figure 8).

Cross-sectional skeletal age data from white Nebraska children of middle socioeconomic level have been reported by Fry (1966).2 z These children were born between 1950 and 1960 and were radiographed at random ages. The sample included 25 children of each sex within each annual interval. The means were below the Greulich-Pyle standards, particularly in 6- and 7-year-old boys. Mareshz 3 reported mixed lon-

United States

Skeletal Maturity of Children

641 years:

Racial, Geographic Area,

and Socioeconomic Differentials

United States

NATIONAL CENTER FOR HEALTH STATISTICS

DIVISION OF HEALTH EXAMINATION STATISTICS

COOPERATION OF THE BUREAU OF THE CENSUS

CONTENTS

0.0

*

SKELETAL MATURITY

iDF CHILDREN 6-11 YEARS:

RACIAL, GEOGRAPHIC AREA, AND SOCIOECONOMIC DIFFERENTIALS

IEzEizz

r

I

-’ z

r

...

““””

.

641 ^years:

-’ ^z